summaryrefslogtreecommitdiff
path: root/results/bci_v2_dev/bci_v2_e0p1_g0p8_c0p03_t22_m0.json
blob: 57f06f95c189ee98976a53ad989c7baae148fde1 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
{
  "args": {
    "critic_eta": 0.03,
    "forward_eta": 0.1,
    "gamma": 0.8,
    "model_seed": 0,
    "outdir": "results/bci_v2_dev",
    "task_seed": 22
  },
  "assays": {
    "challenge": {
      "active_state_episode_steps_by_mode": {
        "acute_critic_lesion": 14336,
        "acute_outcome_lesion": 14336,
        "intact": 14336
      },
      "episodes_per_horizon": 128,
      "horizons": [
        4,
        8,
        12,
        16,
        20,
        24,
        28
      ],
      "selection_over_horizons": false,
      "trajectory_seeds": {
        "12": 432002,
        "16": 432003,
        "20": 432004,
        "24": 432005,
        "28": 432006,
        "4": 432000,
        "8": 432001
      }
    },
    "performance_evaluation_episodes": 256,
    "performance_evaluation_seed": 400022
  },
  "conditions": {
    "critic_training_lesion": {
      "cost": {
        "active_state_episode_steps": 25088,
        "cursor_scalar_observations": 12544,
        "maximum_state_episode_steps": 25088,
        "reverse_mode_calls": 0,
        "role_probe_examples": 6272,
        "task_loss_queries": 0,
        "terminal_outcome_observations": 896
      },
      "critic_l2_after_training": 0.0,
      "daily_success": [
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0
      ],
      "early_success": 0.0,
      "evaluation_active_state_episode_steps": 7168,
      "final_success": 0.0,
      "late_success": 0.0,
      "learning_gain": 0.0,
      "role_cosine_after_training": 0.9946003556251526,
      "training_wall_s": 0.21681366860866547
    },
    "fixed_role": {
      "cost": {
        "active_state_episode_steps": 25088,
        "cursor_scalar_observations": 12544,
        "maximum_state_episode_steps": 25088,
        "reverse_mode_calls": 0,
        "role_probe_examples": 6272,
        "task_loss_queries": 0,
        "terminal_outcome_observations": 896
      },
      "critic_l2_after_training": 0.015532732009887695,
      "daily_success": [
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0
      ],
      "early_success": 0.0,
      "evaluation_active_state_episode_steps": 7168,
      "final_success": 0.0,
      "late_success": 0.0,
      "learning_gain": 0.0,
      "role_cosine_after_training": -0.1393236517906189,
      "training_wall_s": 0.2341603972017765
    },
    "intact": {
      "cost": {
        "active_state_episode_steps": 25088,
        "cursor_scalar_observations": 12544,
        "maximum_state_episode_steps": 25088,
        "reverse_mode_calls": 0,
        "role_probe_examples": 6272,
        "task_loss_queries": 0,
        "terminal_outcome_observations": 896
      },
      "critic_l2_after_training": 0.016242388635873795,
      "daily_success": [
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0
      ],
      "early_success": 0.0,
      "evaluation_active_state_episode_steps": 7168,
      "final_success": 0.0,
      "late_success": 0.0,
      "learning_gain": 0.0,
      "role_cosine_after_training": 0.9946002960205078,
      "training_wall_s": 0.3125084228813648
    },
    "oracle_role": {
      "cost": {
        "active_state_episode_steps": 25088,
        "cursor_scalar_observations": 12544,
        "maximum_state_episode_steps": 25088,
        "reverse_mode_calls": 0,
        "role_probe_examples": 6272,
        "task_loss_queries": 0,
        "terminal_outcome_observations": 896
      },
      "critic_l2_after_training": 0.01627274975180626,
      "daily_success": [
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0
      ],
      "early_success": 0.0,
      "evaluation_active_state_episode_steps": 7168,
      "final_success": 0.0,
      "late_success": 0.0,
      "learning_gain": 0.0,
      "role_cosine_after_training": 1.0000001192092896,
      "training_wall_s": 0.23344024643301964
    },
    "outcome_training_lesion": {
      "cost": {
        "active_state_episode_steps": 25088,
        "cursor_scalar_observations": 12544,
        "maximum_state_episode_steps": 25088,
        "reverse_mode_calls": 0,
        "role_probe_examples": 6272,
        "task_loss_queries": 0,
        "terminal_outcome_observations": 896
      },
      "critic_l2_after_training": 0.016242388635873795,
      "daily_success": [
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0
      ],
      "early_success": 0.0,
      "evaluation_active_state_episode_steps": 7168,
      "final_success": 0.0,
      "late_success": 0.0,
      "learning_gain": 0.0,
      "role_cosine_after_training": 0.9946002960205078,
      "training_wall_s": 0.23566758260130882
    },
    "plasticity_lesion": {
      "cost": {
        "active_state_episode_steps": 25088,
        "cursor_scalar_observations": 12544,
        "maximum_state_episode_steps": 25088,
        "reverse_mode_calls": 0,
        "role_probe_examples": 6272,
        "task_loss_queries": 0,
        "terminal_outcome_observations": 896
      },
      "critic_l2_after_training": 0.015574400313198566,
      "daily_success": [
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0
      ],
      "early_success": 0.0,
      "evaluation_active_state_episode_steps": 7168,
      "final_success": 0.0,
      "late_success": 0.0,
      "learning_gain": 0.0,
      "role_cosine_after_training": 0.9946002960205078,
      "training_wall_s": 0.2053239606320858
    }
  },
  "config": {
    "context_ar": 0.8,
    "context_dim": 16,
    "coupling_scale": 1.0,
    "critic_eta": 0.03,
    "days": 14,
    "eligibility_decay": 0.8,
    "episodes_per_day": 64,
    "feedback": "performance_velocity",
    "forward_eta": 0.1,
    "gamma": 0.8,
    "inertia": 0.65,
    "kappa": 0.0,
    "n_background": 30,
    "n_minus": 5,
    "n_plus": 5,
    "perturb_every": 4,
    "perturb_sigma": 0.03,
    "predictor_eta": 0.2,
    "process_noise": 0.12,
    "steps_per_episode": 28,
    "target": 0.8,
    "terminal_reward": 1.0,
    "vectorizer_eta": 0.03,
    "velocity_reward_scale": 0.25
  },
  "finite": true,
  "hardware": {
    "device": "cpu",
    "platform": "Linux-5.15.0-161-generic-x86_64-with-glibc2.35",
    "threads": 1,
    "torch_version": "2.10.0+cu128"
  },
  "peak_rss_mib": 958.10546875,
  "protocol": {
    "confirmation_seeds_touched": false,
    "d4_gate_sha256": "636c587e47287338cdca7d9558dc3bfb99a2cec615badb23337025d85b8128ef",
    "grid_size": 24,
    "name": "oral_b_v2_development_v1",
    "old_r2_gate_sha256": "4f6f969ceae88afa2523e3472a3373522991f5ecaab69440830c07479d2d3597",
    "protocol_sha256": "940bac74b527605d33a36a2b79f404a69de497a07d7657e5341ad28c4f1373f7",
    "selection_split": "development"
  },
  "provenance": {
    "git_commit": "670c1399659f31ac79c4b454147417d4f8ddca7d",
    "git_tracked_dirty": false,
    "input_sha256": {
      "base_dynamics": "d5a373314562af0daedf55baa04b975fe643236c7821be1560d2cfb4359688fc",
      "confirmation_analyzer": "e18d2f0cee6ee3be74fdf309b51ab80d114206c517918caa501e24379e2046a6",
      "confirmation_runner": "77ee8fa3923481789dba5a864228dd05ef4cc5039ce7f9b4876c3aba477a4939",
      "d4_gate": "636c587e47287338cdca7d9558dc3bfb99a2cec615badb23337025d85b8128ef",
      "development_analyzer": "f5fd61493e2264d4cf39f6915a52a265d932703483a6cb6b063228c2c1ee190d",
      "old_r2_gate": "4f6f969ceae88afa2523e3472a3373522991f5ecaab69440830c07479d2d3597",
      "protocol": "940bac74b527605d33a36a2b79f404a69de497a07d7657e5341ad28c4f1373f7",
      "runner": "4157a57806dedfc2eb70ec84f7a7ca1f73ab684e5d8f0b227cd80465e8e97aa5",
      "v2_dynamics": "786c5aed3091644272ae0a740913ac3cc38cdb02a8cacb6dfc8496e36836cdae",
      "v2_metrics": "421da9287b92e62c742f9eeefbfdccba582c6a4ad5cf92ade4709689120fa1ee"
    },
    "tracked_inputs": {
      "base_dynamics": true,
      "confirmation_analyzer": true,
      "confirmation_runner": true,
      "d4_gate": true,
      "development_analyzer": true,
      "old_r2_gate": true,
      "protocol": true,
      "runner": true,
      "v2_dynamics": true,
      "v2_metrics": true
    }
  },
  "schema_version": 2,
  "signatures": {
    "acute_outcome_lesion_outcome_balanced_acc": 0.5,
    "acute_outcome_lesion_role_aligned_separation": -0.0006112609351708791,
    "causal_role_sign_inversion_index": 0.007273438124456074,
    "challenge_episodes": 896,
    "challenge_failure_count": 896,
    "challenge_success_count": 0,
    "challenge_success_fraction": 0.0,
    "critic_contribution_value_prediction_corr": 0.9999999999966643,
    "decoder_distance_residual_corr": 0.25635489692366603,
    "horizon_success_fraction": {
      "12": 0.0,
      "16": 0.0,
      "20": 0.0,
      "24": 0.0,
      "28": 0.0,
      "4": 0.0,
      "8": 0.0
    },
    "mean_abs_raw_soma_corr": 0.999969695785536,
    "mean_abs_residual_soma_corr": 0.04065599750786667,
    "mean_critic_expectedness_contribution": -0.00041896542624211125,
    "nonterminal_training_events": 24192,
    "raw_minus_residual_abs_soma_corr": 0.9593136982776693,
    "role_aligned_error_cv_corr": 0.3120558519374075,
    "role_aligned_velocity_cv_corr": 0.9988698428175757,
    "surrounding_event_decoder_balanced_acc": 0.5819735166629558,
    "terminal_outcome_separation_drop_under_acute_lesion": 0.0,
    "terminal_previous_soma_outcome_balanced_acc": 0.5,
    "terminal_residual_minus_previous_soma_acc": 0.0,
    "terminal_residual_outcome_balanced_acc": 0.5,
    "terminal_role_aligned_outcome_separation": -0.0006112609351708791,
    "terminal_training_events": 896,
    "velocity_minus_error_abs_cv_corr": 0.6868139908801683
  },
  "split": "development",
  "wall_s": 2.920739535242319,
  "warmup": {
    "critic_training_lesion": {
      "batches": 100,
      "examples": 6400,
      "instruction_present": false,
      "predictor_max_abs_error": 2.384185791015625e-07,
      "rng_seed": 602200,
      "role_cosine_after_warmup": 0.9940935969352722,
      "role_cursor_scalar_observations": 12800,
      "role_update_enabled": true
    },
    "fixed_role": {
      "batches": 100,
      "examples": 6400,
      "instruction_present": false,
      "predictor_max_abs_error": 2.384185791015625e-07,
      "rng_seed": 602200,
      "role_cosine_after_warmup": -0.1393236517906189,
      "role_cursor_scalar_observations": 12800,
      "role_update_enabled": false
    },
    "intact": {
      "batches": 100,
      "examples": 6400,
      "instruction_present": false,
      "predictor_max_abs_error": 2.384185791015625e-07,
      "rng_seed": 602200,
      "role_cosine_after_warmup": 0.9940935969352722,
      "role_cursor_scalar_observations": 12800,
      "role_update_enabled": true
    },
    "oracle_role": {
      "batches": 100,
      "examples": 6400,
      "instruction_present": false,
      "predictor_max_abs_error": 2.384185791015625e-07,
      "rng_seed": 602200,
      "role_cosine_after_warmup": 1.0000001192092896,
      "role_cursor_scalar_observations": 12800,
      "role_update_enabled": false
    },
    "outcome_training_lesion": {
      "batches": 100,
      "examples": 6400,
      "instruction_present": false,
      "predictor_max_abs_error": 2.384185791015625e-07,
      "rng_seed": 602200,
      "role_cosine_after_warmup": 0.9940935969352722,
      "role_cursor_scalar_observations": 12800,
      "role_update_enabled": true
    },
    "plasticity_lesion": {
      "batches": 100,
      "examples": 6400,
      "instruction_present": false,
      "predictor_max_abs_error": 2.384185791015625e-07,
      "rng_seed": 602200,
      "role_cosine_after_warmup": 0.9940935969352722,
      "role_cursor_scalar_observations": 12800,
      "role_update_enabled": true
    }
  }
}